Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Dorsolateral prefrontal cortex supports lexical selection during spoken word recognition
Poster Session D, Saturday, October 26, 10:30 am - 12:00 pm, Great Hall 3 and 4
Mel Mallard1, Chad S. Rogers2, Michael S. Jones1, Sarah McConkey1, Brent Spehar1, Kristin J. Van Engen1, Mitchell S. Sommers1, Jonathan E. Peelle3; 1Washington University in St. Louis, 2Union College, 3Northeastern University
Listening to speech often feels simple and automatic, but there are many factors that can make understanding speech more difficult, including both the content of the speech (for example, psycholinguistic properties) and the acoustic clarity with which it is presented. Although most studies of single word processing have implicated bilateral superior temporal gyri, it has been suggested that regions of the frontal lobe—particularly dorsolateral prefrontal cortex (DLPFC)—are engaged to deal with various types of acoustic or linguistic challenge during listening. Here we tested a hypothesis that left DLPFC is more active in conditions of higher lexical competition in the context of spoken word recognition. We conducted an fMRI study in which we presented words in background noise to both young and older adults. Words varied in lexical competition, being selected to have either few neighbors (i.e., from a sparse phonological neighborhood) or many neighbors (i.e., from a dense phonological neighborhood). Participants were both younger (n = 47; aged 19-30 years) and older (n = 31; aged 65-81 years) adults. Participants performed a speech-repetition task. To measure repetition accuracy, each trial was graded as either a correct or incorrect repetition. Accuracy was generally good, and we restricted further analyses to correct responses only. As expected, at the whole brain level, spoken word recognition in noise was associated with activity in the left and right superior temporal gyrus. To test our hypothesis about DLPFC involvement in spoken word recognition we extracted parameter estimates from each participant, showing a significant density effect [t(77) = 3.36, p < .001, d = 0.38], with Dense words eliciting more activity than Sparse words. Notably, this density effect did not significantly differ by age [t(73.98) = 1.18, p = .241, d = 0.26) and was not significantly correlated with hearing ability (better-ear pure-tone-average) in either the young [r(45) = -.056] or older [r(29) = -.080] group. We conclude that DLPFC, anatomically situated outside classic language regions, participates in lexical selection, being sensitive to the number of competitors a target word has. We found in this region a significant density effect that is robust against age and hearing, revealing a degree of stability in the neural systems underlying spoken word recognition outside the temporal lobe.
Topic Areas: Speech Perception, Control, Selection, and Executive Processes